Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/newstube.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..aes import aes_cbc_decrypt
   9 from ..utils import (
  10     bytes_to_intlist,
  11     int_or_none,
  12     intlist_to_bytes,
  13     parse_codecs,
  14     parse_duration,
  15 )
  16
  17
  18 class NewstubeIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  20     _TEST = {
  21         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
  22         'md5': '9d10320ad473444352f72f746ccb8b8c',
  23         'info_dict': {
  24             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
  25             'ext': 'mp4',
  26             'title': 'Телеканал CNN переместил город Славянск в Крым',
  27             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
  28             'duration': 31.05,
  29         },
  30     }
  31
  32     def _real_extract(self, url):
  33         video_id = self._match_id(url)
  34
  35         page = self._download_webpage(url, video_id)
  36         title = self._html_search_meta(['og:title', 'twitter:title'], page, fatal=True)
  37
  38         video_guid = self._html_search_regex(
  39             r'<meta\s+property="og:video(?::(?:(?:secure_)?url|iframe))?"\s+content="https?://(?:www\.)?newstube\.ru/embed/(?P<guid>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  40             page, 'video GUID')
  41
  42         enc_data = base64.b64decode(self._download_webpage(
  43             'https://www.newstube.ru/embed/api/player/getsources2',
  44             video_guid, query={
  45                 'guid': video_guid,
  46                 'ff': 3,
  47             }))
  48         key = hashlib.pbkdf2_hmac(
  49             'sha1', video_guid.replace('-', '').encode(), enc_data[:16], 1)[:16]
  50         dec_data = aes_cbc_decrypt(
  51             bytes_to_intlist(enc_data[32:]), bytes_to_intlist(key),
  52             bytes_to_intlist(enc_data[16:32]))
  53         sources = self._parse_json(intlist_to_bytes(dec_data[:-dec_data[-1]]), video_guid)
  54
  55         formats = []
  56         for source in sources:
  57             source_url = source.get('Src')
  58             if not source_url:
  59                 continue
  60             height = int_or_none(source.get('Height'))
  61             f = {
  62                 'format_id': 'http' + ('-%dp' % height if height else ''),
  63                 'url': source_url,
  64                 'width': int_or_none(source.get('Width')),
  65                 'height': height,
  66             }
  67             source_type = source.get('Type')
  68             if source_type:
  69                 f.update(parse_codecs(self._search_regex(
  70                     r'codecs="([^"]+)"', source_type, 'codecs', fatal=False)))
  71             formats.append(f)
  72
  73         self._check_formats(formats, video_guid)
  74         self._sort_formats(formats)
  75
  76         return {
  77             'id': video_guid,
  78             'title': title,
  79             'description': self._html_search_meta(['description', 'og:description'], page),
  80             'thumbnail': self._html_search_meta(['og:image:secure_url', 'og:image', 'twitter:image'], page),
  81             'duration': parse_duration(self._html_search_meta('duration', page)),
  82             'formats': formats,
  83         }